/SimpleRL

An implementation of very basic RL algorithms

Primary LanguagePython

SimpleRL

An implementation of basic RL algorithms

This repo only include simple algorithms on simple environments

Current algorithms:

  • Offline Lambda-return (grid world)
  • REINFORCE (CartPole-v1)
  • Actor Critic (Cartpole-v1)
  • Q-learning (Taxi-v3, FrozenLake-v1)
  • Deep Q-Network (CartPole-v1)
  • Priortized Experience Replay (PER) (Breakout-v5)
  • N-step SARSA (FrozenLake-v1)
  • SARSA (FrozenLake-v1)